Focus on... Sophie Abby and the evolution of enzymes' regio-selectivity

on the July 15, 2019

Sophie Abby, a young researcher at TIMC-IMAG, CNRS and Université Grenoble Alpes, is involved in the project “Deciphering the origins of enzymes substrate specificity using large-scale sequence analyses”. This interdisciplinary project received funding from the Grenoble Alpes Data Institute.
Ubiquinone, or co-enzyme Q is a key molecule in the respiratory chain. It is involved in the production of cellular energy in some bacteria, and in eukaryotes (fungi, animals, plants…). Several reactions lead to the synthesis of this lipid, such as three hydroxylation reactions at three different positions. For some species, one enzyme catalyzes these hydroxylation reactions. For others, it was recently shown that two or three enzymes are involved (for Escherichia coli, 3 enzymes: Ubil, H and F, see Figure).


Regio-selectivity variations in the ubiquinone (UQ) biosynthesis pathway. To produce ubiquinone, there are three hydroxylation reactions (addition of an -OH group) are required: on Carbon 5, 1 and 6 of the precursor’s ring. In Escherichia coli, three enzymes from the same family are involved, whereas in Rhodospirullum rubrum, two enzymes are needed, and only one in Neisseria meningitidis. Figure adapted from Pelosi et al. 2016 mSystems (

Therefore the “regio-selectivity” of these  enzymes, that is their capacity to chemically modify particular positions of a given substrate, is either limited (one position can be handled), or broad (several positions are handled by a single enzyme). Interestingly, the enzymes involved in the three hydroxylation reactions of the ubiquinone precursor are part of the same family, that is, they derived from a common ancestral enzyme. Thus, studying these enzymes has the potential to teach us how regio-selectivity evolves and in a broader sense, how enzymes get new functions along evolution.

Sophie Abby in an expert in comparative genomics, phylogenetics and bioinformatics. She is working on this project with Fabien Pierrel and Ludovic Pelosi, biochemists, and Ivan Junier, a biophysicist, all members of the GEM team at the TIMC-IMAG lab. They used comparative genomics to predict the regio-selectivity of this type of enzymes based on their distribution in genomes. They work on more than 1000 bacterial genomes and thus deal with a lot of data to extract the evolutionary history of these enzymes and their associated regio-selectivity. They are now validating their predictions experimentally using complementation assays in mutants from the model organism Escherichia coli.
Thanks to the Grenoble Alpes Data Institute’s funding, two interns were hired. Clothilde Chenal, a master 1 student in bioinformatics (U. Montpellier) was in charge of computing enzymes’ regio-selectivity predictions while Katayoun Kazemzadeh, a master 2 student in biochemistry and genetics (U. Grenoble Alpes), worked on the in vivo validation of the predictions.

This study has the potential to start addressing important questions in biochemistry, structural biology, and evolution: how regio-selectivity evolves, how enzymes operate, etc. But it is also naturally inclined to lead to biotechnology applications and synthetic biology approaches: how can we genetically modify an enzyme, so that it catalyzes the reactions we want?

A first article has just appeared in mBio ( and the team has just obtained fundings for this project, including for a PhD student.

The GEM (Genomics and Evolution of Microorganisms) team at TIMC-IMAG:

Published on January 9, 2020